2006 AIChE Annual Meeting
(633d) Scale-up and Control of Fluidized Bed Production of Solar-Grade Silicon
Authors
This work focuses on developing models for scale-up and process control design of fluidized bed silicon production. Experiments demonstrate that productive operation of the reactor depends on the silicon particle size distribution among other process conditions. We model the silicon particle size distribution with a discrete version of the population balance equation derived from particle mass and number balances. The technique is validated with experimental results.
We apply inventory control theory developed from passivity theory (Farschmann et al. 1998) to the process model. Treating the mass hold-up of particles of a specified size range as controlled inventories and particle flow rates to and from the system as manipulated variables, we are able to perform closed loop simulations to examine sensitivity and prescribe operating conditions. Process sensitivity to varying the hold-up of small particles, the total hold-up, temperature, and the distribution of particles flowing into the reactor is shown. These results are compared with experimental results, and we use the information to determine size control requirements for productive operation of a commercial fluidized bed reactor.
White, C. M., Ege, P., & Ydstie, B. E. (2006), Size distribution modeling for fluidized bed solar-grade silicon production,' Powder Technology 163(2006), 51-58.
Farschman, C. A., Viswanath, K. P. & Ydstie, B. E. (1998), Process systems and inventory control', AIChE Journal 44(8), 1841-1857.